The invention provides a method of introducing two or more genes, involved in lignin biosynthesis, into plant cells. The method of the invention employs either an Agrobacterium-mediated or other appropriate plant gene delivery system by which multiple genes together with a single selectable marker gene are simultaneously transferred and inserted into the genome of plants with high frequencies.
The ability to introduce foreign genes into plants is a prerequisite for engineering agronomic traits in plants. Many systems have been developed for introducing a foreign gene into plant cells, which involve mainly either Agrobacterium- or microprojectile bombardment-mediated transformation (Christou, 1996). The principle of all these systems involves the insertion of a target gene into the host plant genome together with a selectable marker gene encoding either antibiotic or herbicide resistance to aid in the selection of transgenic cells from non-transgenic cells. These systems generally are only effective for introducing a single target gene into the host plant.
To alter agronomic traits, which generally are polygenic in nature, multiple genes involved in complex biosynthetic pathways must be introduced and expressed in plant cells. In this context, the traditional single-gene transfer systems are essentially useless for the following two reasons: 1) it is impractical to introduce multiple genes by repetitive insertion of single genes into transgenic plants due to the time and effort required for recovery of the transgenic tissues; in particular, a repetitive single-gene approach is highly impractical for plant species such as trees which, depending upon the species, require two to three years for transgenic tissue selection and regeneration into a tree; and 2) the presence of a selectable marker gene in a transgenic line precludes the use of the same marker gene in subsequent transformations of plant material from that line. Moreover, the number of available marker genes is limited, and many plant species are recalcitrant to regeneration unless appropriate antibiotic or herbicide selection is used.
Chen et al. (1998) recently reported the genetic transformation of rice with multiple genes by cobombardment of several gene constructs into embryogenic suspension tissues. However, particle bombardment-mediated gene transfer into embryogenic tissues is highly species-dependent, and regeneration of whole plants from embryogenic cells cannot be achieved for a variety of plant species (Horsch et al., 1985).
In contrast, Agrobacterium-mediated gene transfer and whole plant regeneration through organogenesis is a simple process and a less species-dependent system than bombardment-mediated transformation and regeneration via embryogenesis. However, the introduction of more than one gene in a single plasmid vector via Agrobacterium may be technically troublesome and limited by the number or the size of the target genes (Chen et al., 1998). For example, Tricoli et al. (1995) reported the transfer of three target genes to squash via Agrobacterium-mediated gene transfer. A binary plasmid vector containing the three target genes was incorporated into an Agrobacterium strain, which was subsequently used to infect the leaf tissue of squash. As only one line was recovered from numerous infected squash tissues that contained all of the target genes, the use of a single binary vector with a number of genes appears to be a highly inefficient method to produce transgenic plants with multiple gene transfers. Therefore, it was commonly accepted that transfer of multiple genes via Agrobacterium-mediated transformation was impractical (Ebinuma et al., 1997), until success of multiple gene transfer via Agrobacterium was first reported in co-pending, commonly owned PCT application, PCT/US/0027704, filed Oct. 6, 2000, entitled “Method of Introducing a Plurality of Genes into Plants” by Chiang et al, incorporated herein by reference. However, homologous tissue-specific preparation of transgenic trees to specifically alter lignin content, increase S/G (syringyl:guaiacyl) lignin ratio and increase cellulose quantity, as compared to an untransformed plant was unsuccessful.
Yet, the altering of lignin content and composition in plants has been a goal of genetically engineered traits in plants. Lignin, a complex phenolic polymer, is a major part of the supportive structure of most woody plants including angiosperm and gymnosperm trees, which, in turn, are the principal sources of fiber for making paper and cellulosic products. Lignin generally constitutes about 25% of the dry weight of the wood, making it the second most abundant organic compound on earth after cellulose. Lignin provides rigidity to wood for which it is well suited due, in part, to its resistance to biochemical degradation.
Despite its importance to plant growth and structure, lignin is nonetheless problematic to post-harvest, cellulose-based wood/crop processing for fiber, chemical, and energy production because it must be removed or degraded from cellulose at great expense. Certain structural constituents of lignin, such as the guaiacyl (G) moiety, promote monomer cross-linkages that increase lignin resistance to degradation (Sarkanen, 1971; Chang and Sarkanen, 1973; Chiang and Funaoka, 1990). In angiosperms, lignin is composed of a mixture of guaiacyl (G) and syringyl (S) monolignols, and can be degraded at considerably less energy and chemical cost than gymnosperm lignin, which consists almost entirely of guaiacyl moieties (Freudenberg, 1965). It has been estimated that, if syringyl lignin could be genetically incorporated into gymnosperm guaiacyl lignin or into angiosperms to increase the syringyl lignin content, the annual saving in processing of such genetically engineered plants as opposed to their wild types would be in the range of $6 to $10 billion in the U.S. alone. Consequently, there has been long-standing incentive to understand the biosynthesis of syringyl monolignol to genetically engineer plants to contain more syringyl lignin, thus, facilitating wood/crop processing (Trotter, 1990; Bugos et al., 1991; Boudet et al., 1995; Hu et al., 1999).
Depending on the use for the plant, genetic engineering of certain traits has been attempted. For some plants, as indicated above, there has been a long-standing incentive to genetically modify lignin and cellulose to decrease lignin and increase cellulose contents. For example, it has been demonstrated that the digestibility of forage crops by ruminants is inversely proportional to lignin content in plants (Buxton and Roussel, 1988, Crop. Sci., 28, 553-558; Jung and Vogel, 1986, J. Anim., Sci., 62, 1703-1712). Therefore, decreased lignin and high cellulose plants are desirable in forage crops to increase their digestibility by ruminants, thereby providing the animal with more nutrients per unit of forage.
In other plants, genetically increasing the S/G ratio of the lignin has been sought. As noted above, lignin in angiosperms is composed of guaiacyl (G) and syringyl (S) monomeric units, whereas gymnosperm lignin consists entirely of G units. The structural characteristics of G units in gymnosperm lignin promote monomer cross-linkages that increase lignin resistance to chemical extraction during wood pulp production. However, the S units present in angiosperm lignin prevent such chemical resistant cross-links. Therefore, without exception, chemical extraction of G lignin in pulping of gymnosperms is more difficult and requires more chemicals, longer reaction times and higher energy levels than the extraction of G-S lignin during pulping of angiosperms (Sarkanen, K. V., 1971, in Lignins: Occurrence, Formation, Structure and Reaction, Sarkanen, K. V. & Ludwig, C. H., eds., Wiley-Interscience, New York; Chang, H. M. and Sarkanen, K. V., 1973, TAPPI, 56:132-136). As a rule, the reaction rate of extracting lignin during wood pulping is directly proportional to the quantity of the S unit in lignin (Chang, H. M. and Sarkanen, K. V., 1973, TAPPI, 56:132-136). Hence, altering lignin into more reactive G-S type in gymnosperms and into high S/G ratio in angiosperms would represent a pivotal opportunity to enhance current pulping and bleaching efficiency and to provide better, more economical, and more environmentally sound utilization of wood.
Recent results have indicated that high S/G ratio may also add further mechanical advantages to plants, balancing the likely loss of sturdiness of plants with severe lignin reduction (Li et al., 2001, Plant Cell, 13:1567-1585). Moreover, a high S/G lignin ratio would also improve the digestibility of forage crops by ruminants (Buxton and Roussel, 1988, Crop. Sci., 28, 553-558; Jung and Vogel, 1986, J. Anim., Sci., 62, 1703-1712).
In some applications, both a high lignin content and high S/G ratio have been sought (i.e., combining these two traits in plants). For example, it has been demonstrated that when lignin is extracted out from wood during chemical pulping, lignin in the pulping liquor is normally used as a fuel source to provide energy to the pulping and bleaching operations. This lignin-associated energy source, which is not necessary for pulp mills using purchased fuel for energy, is essential to some pulp mills which depend upon internal sources, such as extracted lignin, to be self-sufficient in energy. Therefore, for this purpose, it may be desirable to increase lignin content in pulpwood species, and at the same time to increase the S/G ratio in these species to facilitate the extraction of more lignin to be used as fuel.
Additionally, for grain production and other non-related purposes, increased lignin content and/or S/G lignin ratio are desirable to provide extra sturdiness in plants to prevent the loss of socially and economically important food crops due to dislodging and due to damage to the aerial parts of the plant.
The plant monolignol biosynthetic pathway is set forth in FIG. 1 and will be explained in more detail hereinbelow. The key lignin control sites in the monolignol biosynthetic pathway are mediated by genes encoding the enzymes 4-coumarate-CoA ligase (4CL) (Lee et al., 1997), coniferyl aldehyde 5-hydroxylase (CAld5H) (Osakabe et al., 1999) and S-adenosyl-L-methionine (SAM)-dependent 5-hydroxyconiferaldehyde O-methyltransferase (AldOMT) (Li et al., 2000), respectively, for the formation of sinapaldehyde (see, FIG. 1). Further, coniferyl alcohol dehydrogenase (CAD) (MacKay et al., 1997) catalyzes the reaction including the substrate coniferaldehyde to coniferyl alcohol. It has recently been discovered that sinapyl alcohol dehydrogenase (SAD) enzymatically converts sinapaldehyde into sinapyl alcohol, the syringyl monolignol, for the biosynthesis of syringyl lignin in plants (see, FIG. 1). See, concurrently filed, commonly owned U.S. non-provisional application entitled “Genetic Engineering of Syringyl-Enriched Lignin in Plants,” incorporated herein by reference. It should be noted that the gene encoding the enzyme sinapyl alcohol dehydrogenase (SAD) represents the last gene that is indispensable for genetic engineering of syringyl lignin in plants.
A summary of the conserved regions contained within the coding sequence of each of the above listed proteins is described below. Because SAD is a recently discovered enzyme in Aspen, sequence alignments with other representative species were unable to be performed.
The protein sequence alignments of plant AldOMTs are shown in FIG. 9. All AldOMTs have three conserved sequence motifs (I, II, and III) which are the binding sites of S-adenosyl-L-methionine (SAM), the co-substrate or methyl donor for the OMT reaction (Ibrahim, 1997, Trends Plant Sci., 2:249-250; Li et al., 1997, Proc. Natl. Acad. Sci. USA, 94:5461-5466; Joshi and Chiang, 1998, Plant Mol. Biol., 37:663-674). These signature sequence motifs and the high sequence homology of these proteins to PtAldOMT attest to their function as an AldOMT specific for converting 5-hydroxyconiferaldehyde into sinapaldehyde (Li et al., 2000, J. Biol. Chem., 275:6537-6545), the content of which is incorporated by reference, herein, in its entirety. This AldOMT, like CAld5H, also operates at the aldehyde level of the plant monolignol biosynthetic pathway.
The protein sequence alignments of plant CADs are shown in FIG. 10. It was recently proven that CADs are actually guaiacyl monolignol pathway specific (Li et al., 2001, Plant Cell, 13:1567-1585). Based on high sequence homology, the alignment program picked up CADs from angiosperms as well as gymnosperms (radiata pine, loblolly pine and spruce) which have only G-lignin. All CADs have the Zn1 binding motif and structural Zn2 consensus region, as well as a NADP binding site (Jornvall et al., 1987, Eur. J. Biochem., 167:195-201; MacKay et al., 1995, Mol. Gen. Genet., 247:537-545). All these sequence characteristics and high sequence homology to PtCAD attest to these CAD function as a G-monolignol specific CAD (Li et al., 2001, Plant Cell, 13:1567-1585).
The protein sequence alignments of plant Cald5Hs are shown in FIG. 11. Although, there are different types of 5-hydroxylases, i.e., F5H, CAld5H is the sole enzyme catalyzing specifically the conversion of coniferaldehyde into 5-hydroxyconiferaldehyde. All full-length CAld5Hs have the proline-rich region located from amino acid 40 to 45 which is believed to be involved in the process of correct folding of microsomal P450s and is also important in heme incorporation into P450s (Yamazaki et al. 1993, J. Biochem. 114:652-657). Also they all have the heme-binding domain (PFGXGXXXCXG) that is conserved in all P450 proteins (Nelson et al. 1996, Pharmacogenetics, 6:1-41). These signature sequences and the high sequence homology of these proteins to PtCAld5H their function as a 5-hydroxylase that is specific for converting coniferaldehyde into 5-hydroxyconiferaldehyde (Osakabe et al., 1999, Proc. Natl. Acad. Sci. USA, 96:8955-8960).
The protein sequence alignment of plant 4CLs are shown in FIG. 12. In general, 4CL catalyzes the activation of the hydroxycinnamic acids to their corresponding hydroxycinnamoyl-CoA esters. 4CL has the highest activity with p-coumaric acid. 4CL cDNA sequences have been reported from a number of representative angiosperms and gymnosperms, revealing two highly conserved regions, a putative AMP-binding region (SSGTTGLPKGV), and a catalytic motif (GEICIRG). The amino acid sequences of 4CL from plants contain a total of five conserved Cys residues.
Despite recognition of these key enzymes in lignin biosynthesis, there continues to be a need to develop an improved method to simultaneously control the lignin quantity, lignin compositions, and cellulose contents in plants by introducing multiple genes into plant cells.